- Email: [email protected]
Androgen receptors are differentially distributed between right and left cerebral hemispheres of the fetal male rhesus monkey
Brain Research, 516 (1990) 122-126 Elsevier
122 BRES 15444
Androgen receptors are differentially distributed between right and left cerebral hemispheres of the fetal male rhesus monkey Samuel A. Sholl and Kil L. Kim Wisconsin Regional Primate Research Center, University of Wisconsin, Madison, WI 53715-1299 (U.S.A.) (Accepted 10 October 1989)
Key words: Androgen receptor; Fetus; Brain; Primate; Cerebral lateralization
In humans there are apparent sex differences in verbal and spatial abilities as well as several cortical pathologies. These differences may arise as the result of prenatal androgen exposure and its effect on the development of the cerebral cortex. With this in mind, we have examined androgen receptor (AR), aromatase (AROM) and 5a-reductase (5aR) levels in the cerebral cortex of Day 70 male and female fetal rhesus monkeys (Macaca mulatta). Receptor and enzyme levels were evaluated in both right (Rt) and left (Lft) temporal (TMP) and frontal (FR) lobes of the cerebral cortex. AR levels in FR-Rt of male subjects were higher than levels in FR-Lft (for each and every subject, P < 0.05), while in females, there was no consistent pattern in the distribution of the receptor between the two sides of FR. In contrast, AR values in TMpoLft of male subjects were consistently higher than in TMP-Rt (P < 0.05). As with the FR, females exhibited no consistent pattern in the distribution of AR between the two TMP sides. AROM and 5aR levels were similar, regardless of sex, between both sides of the two cortical lobes indicating that the AR distribution pattern is not a general biochemical phenomenon, The differential cortical distribution of AR in fetal males versus females lends support to the hypothesis that prenatal androgens from the fetal testes may effect the differentiation of sexually dimorphic, side-specific cortical activity. INTRODUCTION Sex differences have been described in humans for both verbal and spatial abilities 13'26. Sex-dependencies have also b e e n n o t e d for several cortical pathologies. Dyslexia and autism occur m o r e frequently in males 22. Taylor and O u n s t e d 23 noted in epileptic patients a significant unilateral distribution of mesial temporal sclerosis resulting from febrile convulsions during infancy. This distribution was d e p e n d e n t upon both the sex of the patient and the age at which convulsions first occurred. B a r r et al. 1 r e p o r t e d a higher incidence of frontal lobe arteriovenous malformations in male than female patients. Taken together, these p h e n o m e n a suggest that early brain d e v e l o p m e n t , including the lateralization of cortical activity, m a y be susceptible to horm o n a l modification. E x p e r i m e n t a l evidence which supports the influence of fetal androgens on brain differentiation comes from studies in which it was d e m o n s t r a t e d that the t r e a t m e n t of pregnant rhesus m o n k e y s with androgens between Days 45 and 100 of gestation increases mounting and 'rough and t u m b l e ' play behavior in female offspring 9'1°. These experiments were designed to provide the female fetus with the same exposure to androgens as is normally found in the male
fetus at this stage of gestation 17. A n d r o g e n receptors ( A R s ) have been found in various regions of the M a c a q u e cerebral cortex 3'11'15'16'21. Receptors have b e e n m e a s u r e d in b o t h males and females at different stages of gestation, and their presence suggests that differentiation of the cortex m a y be susceptible to androgen modification. These studies, however, did not focus on right versus left side differences in r e c e p t o r levels. Such differences might be expected in light of the sex-dependent differences in cortical activities and pathologies n o t e d in humans. W i t h this in mind, we have measured androgen r e c e p t o r levels in the rhesus m o n k e y cerebral cortex at a time when in u t e r o androgen t r e a t m e n t has been found to be effective in modifying behavioral patterns of female offspring 9"1°. A R levels were assessed in both the right and left frontal and t e m p o r a l lobes of the cerebrum. A r o m a t a s e and 5areductase activities were also evaluated, since both enzymes m a y be i m p o r t a n t in determining the overall efficacy of androgen exposure.
MATERIALS AND METHODS Five male and six female fetal rhesus monkeys (Macaca mulatta) produced by timed matings at the Wisconsin Regional Primate
Correspondence: S.A. Sholl, Wisconsin Regional Primate Research Center, University of Wisconsin, 1223 Capitol Court, Madison, WI 53715-1299, U.S.A. 0006-8993/90/$03.50 (~) 1990 Elsevier Science Publishers B.V. (Biomedical Division)
123 Research Center were used in the present study. The gestation period of rhesus monkeys at this facility is approximately 166 days. Fetuses were taken at 70 days postconception, at which time mothers were sedated with ketamine hydrochloride (15 mg/kg) and atropine sulfate (0.4 mg) and anesthetized through an endotrachial tube with nitrous oxide-oxygen (4 l/min-2 1/min). Fetuses were delivered by cesarean section.
FRONTAL LOBE ANDROGEN RECEPTOR Male
Femalo
6,
._.R
5.
P ~" ot
q
4..
I
I
n
o
Preparation of tissue Immediately after removing the brain, coronal sections were cut from right (Rt) and left (Lft) temporal (Tmp) and frontal (F) lobe regions (one section per region). These sections were approximately 1-2 mm thick and consisted mostly of grey matter which was distinguishable from the underlying more gelatinous white matter. Within each sex, average weights of cortical samples were similar between sides (male: 138 _+ 25, 161 + 34, 116 _+ 12, 102 _+ 16; female: 102 + 7, 94 + 14, 95 _+ 7, 97 + 10 mg - - for F-Rt, F-Lft, Tmp-Rt and Tmp-Lft, respectively). To provide a basis for assessing the relative importance of enzyme levels, Leg skeletal muscle was also taken. Preparative techniques used to quantify receptor and steroid enzyme levels in a single tissue sample have been described previously2°. Briefly, tissue was placed in cold buffer (Buffer-A) consisting of 10 mM Tris-HCl, 1.5 mM EDTA, 1 mM mercaptoethanol, 25 mM sodium molybdate and 10% (v/v) glycerol (pH 7.4). The tissue was rinsed once with 1 mi Buffer-A. It was then homogenized in Buffer-A (100-200 mgeq/ml buffer) using a Brinkmann Polytron homogenizer (Westbury, NY) (three 3-5 s bursts, on ice). A portion of this homogenate was used immediately for the analysis of aromatase and 5a-reductase activities. The remaining homogenate was centrifuged at 2000 g for 15 rain and the resultant supernatant removed. The pellet was washed with 1 ml of Buffer-A and the centrifugation repeated. The resultant and previous supernatants were combined and centrifuged at 105,000 g to yield a cytosolic fraction. This fraction was frozen at -80 °C and analyzed for androgen receptor activity 2-3 months later.
Analysis of cytosolic androgen receptor The analysis of androgen receptor levels in the fetal cerebral cortex, as carried out in the current study, has been validated and described previously]6. Estimation of receptor activity involved the 24-h incubation of cytosol at 4°C with 4 nM [1,2,4,5,6, 7-3H]dihydrotestosterone (119 Ci/mmol, DuPont NEN) in the presence or absence of 1 ~M methyltrienolone (R1881, DuPont NEN). Following the incubation, free and bound ligand were separated by gel filtration on Sephadex LH-20 columns. The level of specific binding was related to the amount of protein in the cytosol. Interassay variation was less than 8%. Samples from both cortical sides of each animal were analysed in the same assay, and determinations were performed in duplicate•
Analysis of aromatase and 5a-reductase activities Both enzymes were measured in tissue homogenate using procedures which have been described previously2°. Briefly, aromatase activity was estimated from the synthesis of 3H20 from [lfl-aH]testosterone (21 Ci/mmol), while 5a-reductase activity was assessed from the formation of [3H]dihydrotestosterone from [1,2-3H]testosterone (55 Ci/mmol, DuPont NEN). Reactions were carried out in the presence of NADPH plus NADH (0.4 mM each) for 6 h at 37 °C under an atmosphere of O2/CO2. Products were isolated by column and thin layer chromatography 2°. Reaction rates were related to the amount of protein in the homogenate.
Protein and radioactivity measurements Protein values were estimated by the method of Bradford 2. Radioactivity was measured using either Triton X-toluene-Omnifluor (DuPont NEN) or toluene-Liquifluor (DuPont NEN) scintillation fluid. Counting time was set to yield a dpm S.D. of less than 5%.
0
,E 2.
Right
Loft
Left
Right
TEMPORAL LOBE ANDROGEN RECEPTOR 12.
Male
Female et
¢
10' .e
~E
•
6.
M 0
E
4.
."
.,..
U_
.
2. Right
Left
Right
Loft
Side Fig. 1. Androgen receptor levels in the frontal and temporal lobes of the cerebral cortex. Dashed lines connect right and left side values in individual male and female subjects. Open bars represent the means of the respective groups•
RESULTS
Androgen receptor concentration A R levels in the right frontal lobe of male subjects were higher than in the left frontal lobe (for each and every subject; A N O V A : P < 0.05; Wilcoxon's signed rank test (one tailed): P < 0.05) (Fig. 1), while in female fetuses, there was no consistent pattern in the distribution of the receptor between the two sides of the frontal lobe. In contrast, A R values in the left temporal lobe of male subjects were consistently higher than in the right temporal lobe (Wilcoxon's signed rank test: P < 0.05). As with the frontal lobe, female fetuses exhibited no consistent pattern in the distribution of A R between the two sides of the temporal lobe. Average cortical A R levels in males were: 3.4 (F-Rt), 0.4 (F-Lft), 1.4 (Tmp-Rt) and 3.5 fmoles/mg protein (Tmp-Lft), while in females, concentrations were 1.2 (F-Rt), 1.4 (F-Lft), 4.5 ( T m p - R t ) and 1.1 fmol/mg protein (Tmp-Lft). Substantial variation was noted between subjects in right minus left hemisphere A R levels. For the frontal
124 ,---
c - . _C "-- G)
• Frontal Cortex o Temporal Cortex
0~3_ i-
5
1•0-
•
5 a - R E D U C T A S E ACTIVITY
4.
........................... • -..... ..-"" " ' " ' " ' " ' "•
hE
~ "'"'"..
• ,.-""
0
0
08. Male Female
. -c
~ o
•
0.6-
rJ
0
WE
2
~
o
-0
E
t3._
<
0 3.8
43
48
518
o s'.3
0.4-
~
0
0.2-
0.0 Mus
CFr
CFI
CYr
CYl
Bruin Weight (gm)
Fig. 2. The relation between brain weight in individual male fetuses and right minus left side androgen receptor levels in the frontal lobe (closed circles) and temporal lobe (open circles) of the cerebral cortex• While temporal lobe AR side differences did not appear to be related to brain weight, frontal lobe AR differences could be correlated to brain weight• This relationship was best described by the polynomial equation, Y = -49.7 + 19.9X-1.84X 2, as determined by least squares regression analysis (rE = 0.857)• The dashed line describes the fitted equation.
lobe of male subjects, this variation a p p e a r e d to be related to brain weight (Fig. 2, second degree polynomial, r e = 0.857, regression F2, 2 = 5.99). H o w e v e r , for the male t e m p o r a l lobe and for both lobes in females (data not shown), there was no a p p a r e n t relationship between the right-left side distribution of A R levels and brain weight. Total male and female brain weights were significantly different ( A N O V A : P -- 0.04) averaging 4.93 + 0.25 g for males and 3.93 + 0.34 g for females•
Enzyme levels T h e r e was no statistical difference in aromatase activity b e t w e e n any of the tissues or sexes which were e x a m i n e d including muscle (Fig. 3). M o r e o v e r , in contrast to A R , specific side differences were not found 1.0-
AROMATASE ACTIVITY ~- 0.8._c
[--]
£ 0.6-
Male Female
o_
E 0.40
E 0.2-
13..
oo
rTJ Mus
CFr
CFI
CTr
CTI
Tissue
Fig. 3. Aromatase activity in muscle and cerebral cortex of male and female fetuses. Activities were measured in skeletal muscle (Mus) and in the frontal and temporal lobes of the cortex on the right (CFr, CTr) and left (CFI, CTI) sides• Data represent the means and S.E.M.
Tissue
Fig. 4.5a-Reductase activity in muscle and cerebral cortex of male and female fetuses. Activities were measured in skeletal muscle (Mus) and in the frontal and temporal lobes of the cortex on the right (CFr, CTr) and left (CFI, CTI) sides• Data represent the means and S.E.M.
within individual subjects ( d a t a not shown). For 5a-reductase activity, A N O V A indicated a significant tissue effect ( P < 0.001) which could be explained by higher enzyme activities in the various cortex regions than in skeletal muscle (Fig. 4). T h e r e was no significant sex, regional, side or individual difference in 5a-reductase activity in the cerebral cortex. DISCUSSION
The current study d e m o n s t r a t e s a unilateral distribution of A R in the frontal and t e m p o r a l lobes of the male cerebral cortex. This is the first r e p o r t of s e x - d e p e n d e n t biochemical lateralization of the developing p r i m a t e brain. The fact that neither a r o m a t a s e n o r 5a-reductase exhibited lateralization indicates that the o b s e r v e d distribution of A R is not a general p h e n o m e n o n . A n d r o g e n s affect the d e v e l o p m e n t of neural systems in other species and in conjunction with the observed unilateral distribution of A R in the m o n k e y cerebrum, may influence cortical d e v e l o p m e n t in primates. Evidence in birds and rodents indicates a direct facilitatory effect of androgens on synaptogenesis and an inhibitory effect on synapse elimination 4'12'27. A n d r o g e n s also stimulate the proliferation of glial and endothelial cells 8, retard neuronal loss resulting from a x o t o m y 28 and promote neurite outgrowth from explants 2s. A l t h o u g h several of these actions may be m e d i a t e d through the aromatization of androgens to estrogens, some may d e p e n d directly upon androgens. In the androgensensitive levator ani muscle of the male rat, synapse elimination occurs m o r e slowly following birth than in the androgen-insensitive extensor digitorum longus muscle 12. Specific antiandrogens also have been shown to block the
125 masculinization of the spinal nucleus of the bulbocavernosus 5. The similarity between muscle and cortical aromatase activities, noted in the current study, as well as the variable distribution or absence of cytosolic estrogen receptors on Day 70 (Sholl, unpublished) strengthen the conclusion that any potential effects of androgens on the primate cerebral cortex do not depend upon their aromatization to estrogens. This conclusion is also supported by the observation that administration of dihydrotestosterone, a non-aromatizable androgen, to pregnant monkeys during this early stage of development results in more masculine behavior in female offspring 9. It is not clear how the side differences in AR levels arise, although the apparent correlation in males between these differences in the frontal lobe and total brain weight suggests that, at least in this region, they may arise from the maturational processes which occur around this time. This is supported by preliminary observations from our laboratory on Day 50 fetuses which revealed little or no lateralization of A R regardless of sex, and in several fetuses, levels were undetectable. Although it is possible that side differences in measured A R levels reflect differences in receptor stability and/or degradation, a more plausible explanation is that the receptor is localized within specific cell types or structures which are themselves unequally distributed. If one hypothesizes that there are early side differences in hemisphere maturation, regardless of sex, then the presence of circulating androgen may accentuate the maturational imbalance through interaction with the AR. This might occur as a result of changes in the rate of synapse formation or elimination. A disproportionate stimulation of these processes on one side, as a consequence of higher AR levels on that side, might lead in turn to a decline on the contralateral side. In this regard, it has been proposed that initially there is a pool of neurons which are competitively trying to establish connections with one another 6. If neurons fail to establish these connections, they die. The unilateral distribution of AR may favor the formation of synapses on the side which exhibits higher A R levels. As a result, neurons on the opposite side, which are trying to establish contralateral connections, would tend to decline in numbers. With this loss would come a diminution in AR-associated activity. In contrast, when circulating androgen levels are low, as in the female fetus, the maturation of right and left cerebral hemispheres would tend to proceed in a more random fashion. Evidence in primates which supports the concept of neuronal competition comes from the work of Goldman et al. 7. After unilateral lesioning of the frontal lobe of a fetal monkey ( - E l l 0 ) , the frontal lobe on the opposite
side not only developed connections to the ipsilateral thalamus and caudate nucleus (normal pattern), but a!so established connections with these structures on the side of the lesion. It has been suggested 6 that the compensatory increase in contralateral connections from the non-lesioned side was due to the fact that the lesion removed competing neurons from the pool of those trying to establish connections. Whatever the cause of the unilateral distribution of AR in the male fetus, it still remains that such a distribution pattern was noted. Moreover, this implies that the degree of androgen action in the cortex is dissimilar between the two sides, offering a possible explanation for the ontogeny of sex differences in side-dependent cortical activities and pathologies. This explanation (at least for the frontal lobe) is in line with the hypothesis of Geschwind and Galaburda 6 that there is an androgen-dependent delay in left hemisphere maturation. 5a-Reductase activity was significantly higher in the various cortical regions than in muscle suggesting its importance in these brain areas. This enzyme catalyzes the formation of dihydrotestosterone from testosterone. Since dihydrotestosterone binds more avidly than testosterone to AR, elevated 5a-reductase levels may serve to enhance side-specific, AR-mediated changes. In a previous study, we noted high levels of 5a-reductase in the fetal monkey corpus callosum 2°. On the basis of this finding, it is tempting to speculate that the enzyme helps to concentrate dihydrotestosterone in the corpus callosum. In so doing, the reductase may facilitate androgendependent receptor activity in those cortical regions receiving caUosal projections, particularly in those cells which contain elevated A R levels. When estradiol is implanted unilaterally into the rat hypothalamus, there are differential behavioral and neuroendocrine effects depending upon the side of exposure 14'~8. In addition, cytosolic estrogen receptor levels in the rat cerebral cortex on postnatal day 2 exhibit a significant right-left asymmetry depending o n s e x 19. Although the behavioral and neuroendocrine importance of these phenomena remain to be established, they do suggest that the rat may be an important model for future studies of biochemical asymmetry and the ontogeny of this asymmetry. Recently, Tobet and Fox 24 using a monoclonal antibody, 'AB-2', noted an asymmetrical distribution of immunoreactivity in the neonatal male rat brain in a region caudal to the optic chiasm. A higher density of immunoreactivity was evident in radial glia on the right side. Although females did not exhibit the same degree of asymmetry, early treatment with testosterone shifted the distribution pattern closer to that found in males. This study not only provides additional evidence
126 of asymmetry in the developing rat brain, but also describes an experimental approach which may be applicable to future investigations into the ontogeny of brain asymmetry in both the rat and primate.
REFERENCES 1 Barr, W.B., Jaffe, J., Wasserstein, J., Michelson, W.J. and Stein, B.M., Regional distribution of cerebral arteriovenous malformations, Arch. Neurol., 46 (1988) 410-412. 2 Bradford, M., A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of proteindye binding, Anal. Biochem., 72 (1976) 248-254. 3 Clark, A.S., MacLusky, N.J. and Goldman-Rakic, P.S., Androgen binding and metabolism in the cerebral cortex of the developing rhesus monkey, Endocrinology, 123 (1988) 932-940. 4 DeVoogd, T.J., Nixdorf, B. and Nottebohm, E, Synaptogenesis and changes in synaptic morphology related to acquisition of a new behavior, Brain Research, 329 (1985) 304-308. 5 Fishman, R.B. and Breedlove, S.M., Androgen blockade of bulbocavernosus muscle inhibits testosterone-dependent masculinization of spinal motoneurons in newborn female rats, Soc. Neurosci. Abstr., 13 (1987) 1520. 6 Geschwind, N. and Galaburda, A.M., Cerebral Lateralization, Biological Mechanisms, Associations, and Pathology, MIT, Cambridge, 1987, 283 pp. 7 Goldman, P.S. and Galkin, T.W., Prenatal removal of frontal association cortex in the fetal rhesus monkey: anatomical and functional consequences in postnatal life, Brain Research, 152 (1978) 451-485. 8 Goldman, S.A. and Nottebohm, E, Neuronal production, migration and differentiation in a vocal control nucleus of the adult female canary brain, Proc. Natl. Acad. Sci. U.S.A., 80 (1983) 2390-2394. 9 Goy, R.W., Differentiation of male social traits in female rhesus macaques by prenatal treatment with androgens: variation in type of androgen, duration, and timing of treatment. In M.J. Novy and J.A. Resko (Eds.), Fetal Endocrinology, Academic Press, New York, 1981, pp. 319-339. 10 Goy, R.W., Uno, H. and Sholl, S.A., Psychological and anatomical consequences of prenatal exposure to androgens in female rhesus. In T. Mori and H. Nagasawa (Eds.), Toxicity of Hormones in Perinatal Life, CRC Press, Boca Raton, 1988, pp. 127-144. 11 Handa, R.J., Connolly, P.B. and Resko, J.A., Ontogeny of cytosolic androgen receptors in the brain of the fetal rhesus monkey, Endocrinology, 122 (1988) 1890-1896. 12 Jordan, C.L., Letinsky, M.S. and Arnold, A.P., Synapse elimination occurs late in the hormone-sensitive levator ani muscle of the rat, J. Neurobiol., 19 (1988) 335-356. 13 Maccoby, E.M. and Jacklin, C.N., The Psychology of Sex Differences, Stanford University Press, Stanford, 1974, 634 pp. 14 Nordeen, E.J. and Yahr, P., Hemispheric asymmetries in the
Acknowledgements. The work described in this article, Publication 29-030 of the Wisconsin Regional Primate Research Center, was supported by NIH Grants HD18865 and RR00167. All experiments presented in this manuscript were performed by following the standards established by the Animal Welfare Act and by documents entitled 'Principles for the Use of Animals' and 'Guide for the Care and Use of Laboratory Animals'.
behavioral and hormonal effects of sexually differentiating mammalian brain, Science, 218 (1982) 391-394. 15 Pomerantz, S.M., Fox, T.O., Sholl, S.A., Vito, C.C. and Goy, R.W., Androgen and estrogen receptors in fetal rhesus monkey brain and anterior pituitary, Endocrinology, 116 (1985) 83-89. 16 Pomerantz, S.M. and Sholl, S.A., Analysis of sex and regional differences in androgen receptors in fetal rhesus monkey brain, Dev. Brain Res., 36 (1987) 151-154. 17 Resko, J.A., Ellinwood, W.E., Pasztor, L.M. and Buhl, A.E., Sex steroids in the umbilical circulation of fetal rhesus monkeys from the time of gonadal differentiation, J. Clin. Endocrinol. Metab., 50 (1980) 900-905. 18 Roy, E.J. and Lynn, E.M., Asymmetry in responsiveness of the hypothalamus of the female rat to estradiol, Physiol. Behav., 40 (1987) 267-269. 19 Sandhu, S., Cook, P. and Diamond, M.C., Rat cerebral cortical estrogen receptors: male-female, right-left, Exp. Neurol., 92 (1986) 186-196. 20 ShoU, S.A., Goy, R.W. and Kim, K.L., 5a-Reductase, aromatase, and androgen receptor levels in the monkey brain during fetal development, Endocrinology, 124 (1989) 627-634. 21 Sholl, S.A. and Pomerantz, S.M., Androgen receptors in the cerebral cortex of fetal female rhesus monkeys, Endocrinology, 119 (1986) 1625-1631. 22 Taylor, D.C., The influence of sexual differentiation on growth, development, and disease. In J.A. Davis and J. Dobbing (Eds.), Scientific Foundations of Paediatrics, Saunders, Philadelphia, 1974, pp. 29-44. 23 Taylor, D.C. and Ounsted, C., Biological mechanisms influencing the outcome of seizures in response to fever, Epilepsia, 12 (1971) 33-45. 24 Tobet, S.A. and Fox, T.O., Sex- and hormone-dependent antigen immunoreactivity in developing rat hypothalamus, Proc. Natl. Acad. Sci. U.S.A., 86 (1989) 382-386. 25 Toran-Allerand, C.D., Sex steroids and the development of the newborn mouse hypothalamus and preoptic area in vitro: implications for sexual differentiation, Brain Research, 106 (1976) 407-412. 26 Wittig, M.A. and Petersen, A.C., Sex-Related Differences in Cognitive Functioning, Academic Press, New York, 1979, 378 PP. 27 Wright, L.L. and Smolen, A.J., Neonatal testosterone treatment increases neuron and synapse numbers in male rat superior cervical ganglion, Dev. Brain Res., 8 (1983) 145-153. 28 Yu, W.A., Sex difference in neuronal loss induced by axotomy in the rat brain stem motor nuclei, Exp. Neurol., 102 (1988) 230-235.
122 BRES 15444
Androgen receptors are differentially distributed between right and left cerebral hemispheres of the fetal male rhesus monkey Samuel A. Sholl and Kil L. Kim Wisconsin Regional Primate Research Center, University of Wisconsin, Madison, WI 53715-1299 (U.S.A.) (Accepted 10 October 1989)
Key words: Androgen receptor; Fetus; Brain; Primate; Cerebral lateralization
In humans there are apparent sex differences in verbal and spatial abilities as well as several cortical pathologies. These differences may arise as the result of prenatal androgen exposure and its effect on the development of the cerebral cortex. With this in mind, we have examined androgen receptor (AR), aromatase (AROM) and 5a-reductase (5aR) levels in the cerebral cortex of Day 70 male and female fetal rhesus monkeys (Macaca mulatta). Receptor and enzyme levels were evaluated in both right (Rt) and left (Lft) temporal (TMP) and frontal (FR) lobes of the cerebral cortex. AR levels in FR-Rt of male subjects were higher than levels in FR-Lft (for each and every subject, P < 0.05), while in females, there was no consistent pattern in the distribution of the receptor between the two sides of FR. In contrast, AR values in TMpoLft of male subjects were consistently higher than in TMP-Rt (P < 0.05). As with the FR, females exhibited no consistent pattern in the distribution of AR between the two TMP sides. AROM and 5aR levels were similar, regardless of sex, between both sides of the two cortical lobes indicating that the AR distribution pattern is not a general biochemical phenomenon, The differential cortical distribution of AR in fetal males versus females lends support to the hypothesis that prenatal androgens from the fetal testes may effect the differentiation of sexually dimorphic, side-specific cortical activity. INTRODUCTION Sex differences have been described in humans for both verbal and spatial abilities 13'26. Sex-dependencies have also b e e n n o t e d for several cortical pathologies. Dyslexia and autism occur m o r e frequently in males 22. Taylor and O u n s t e d 23 noted in epileptic patients a significant unilateral distribution of mesial temporal sclerosis resulting from febrile convulsions during infancy. This distribution was d e p e n d e n t upon both the sex of the patient and the age at which convulsions first occurred. B a r r et al. 1 r e p o r t e d a higher incidence of frontal lobe arteriovenous malformations in male than female patients. Taken together, these p h e n o m e n a suggest that early brain d e v e l o p m e n t , including the lateralization of cortical activity, m a y be susceptible to horm o n a l modification. E x p e r i m e n t a l evidence which supports the influence of fetal androgens on brain differentiation comes from studies in which it was d e m o n s t r a t e d that the t r e a t m e n t of pregnant rhesus m o n k e y s with androgens between Days 45 and 100 of gestation increases mounting and 'rough and t u m b l e ' play behavior in female offspring 9'1°. These experiments were designed to provide the female fetus with the same exposure to androgens as is normally found in the male
fetus at this stage of gestation 17. A n d r o g e n receptors ( A R s ) have been found in various regions of the M a c a q u e cerebral cortex 3'11'15'16'21. Receptors have b e e n m e a s u r e d in b o t h males and females at different stages of gestation, and their presence suggests that differentiation of the cortex m a y be susceptible to androgen modification. These studies, however, did not focus on right versus left side differences in r e c e p t o r levels. Such differences might be expected in light of the sex-dependent differences in cortical activities and pathologies n o t e d in humans. W i t h this in mind, we have measured androgen r e c e p t o r levels in the rhesus m o n k e y cerebral cortex at a time when in u t e r o androgen t r e a t m e n t has been found to be effective in modifying behavioral patterns of female offspring 9"1°. A R levels were assessed in both the right and left frontal and t e m p o r a l lobes of the cerebrum. A r o m a t a s e and 5areductase activities were also evaluated, since both enzymes m a y be i m p o r t a n t in determining the overall efficacy of androgen exposure.
MATERIALS AND METHODS Five male and six female fetal rhesus monkeys (Macaca mulatta) produced by timed matings at the Wisconsin Regional Primate
Correspondence: S.A. Sholl, Wisconsin Regional Primate Research Center, University of Wisconsin, 1223 Capitol Court, Madison, WI 53715-1299, U.S.A. 0006-8993/90/$03.50 (~) 1990 Elsevier Science Publishers B.V. (Biomedical Division)
123 Research Center were used in the present study. The gestation period of rhesus monkeys at this facility is approximately 166 days. Fetuses were taken at 70 days postconception, at which time mothers were sedated with ketamine hydrochloride (15 mg/kg) and atropine sulfate (0.4 mg) and anesthetized through an endotrachial tube with nitrous oxide-oxygen (4 l/min-2 1/min). Fetuses were delivered by cesarean section.
FRONTAL LOBE ANDROGEN RECEPTOR Male
Femalo
6,
._.R
5.
P ~" ot
q
4..
I
I
n
o
Preparation of tissue Immediately after removing the brain, coronal sections were cut from right (Rt) and left (Lft) temporal (Tmp) and frontal (F) lobe regions (one section per region). These sections were approximately 1-2 mm thick and consisted mostly of grey matter which was distinguishable from the underlying more gelatinous white matter. Within each sex, average weights of cortical samples were similar between sides (male: 138 _+ 25, 161 + 34, 116 _+ 12, 102 _+ 16; female: 102 + 7, 94 + 14, 95 _+ 7, 97 + 10 mg - - for F-Rt, F-Lft, Tmp-Rt and Tmp-Lft, respectively). To provide a basis for assessing the relative importance of enzyme levels, Leg skeletal muscle was also taken. Preparative techniques used to quantify receptor and steroid enzyme levels in a single tissue sample have been described previously2°. Briefly, tissue was placed in cold buffer (Buffer-A) consisting of 10 mM Tris-HCl, 1.5 mM EDTA, 1 mM mercaptoethanol, 25 mM sodium molybdate and 10% (v/v) glycerol (pH 7.4). The tissue was rinsed once with 1 mi Buffer-A. It was then homogenized in Buffer-A (100-200 mgeq/ml buffer) using a Brinkmann Polytron homogenizer (Westbury, NY) (three 3-5 s bursts, on ice). A portion of this homogenate was used immediately for the analysis of aromatase and 5a-reductase activities. The remaining homogenate was centrifuged at 2000 g for 15 rain and the resultant supernatant removed. The pellet was washed with 1 ml of Buffer-A and the centrifugation repeated. The resultant and previous supernatants were combined and centrifuged at 105,000 g to yield a cytosolic fraction. This fraction was frozen at -80 °C and analyzed for androgen receptor activity 2-3 months later.
Analysis of cytosolic androgen receptor The analysis of androgen receptor levels in the fetal cerebral cortex, as carried out in the current study, has been validated and described previously]6. Estimation of receptor activity involved the 24-h incubation of cytosol at 4°C with 4 nM [1,2,4,5,6, 7-3H]dihydrotestosterone (119 Ci/mmol, DuPont NEN) in the presence or absence of 1 ~M methyltrienolone (R1881, DuPont NEN). Following the incubation, free and bound ligand were separated by gel filtration on Sephadex LH-20 columns. The level of specific binding was related to the amount of protein in the cytosol. Interassay variation was less than 8%. Samples from both cortical sides of each animal were analysed in the same assay, and determinations were performed in duplicate•
Analysis of aromatase and 5a-reductase activities Both enzymes were measured in tissue homogenate using procedures which have been described previously2°. Briefly, aromatase activity was estimated from the synthesis of 3H20 from [lfl-aH]testosterone (21 Ci/mmol), while 5a-reductase activity was assessed from the formation of [3H]dihydrotestosterone from [1,2-3H]testosterone (55 Ci/mmol, DuPont NEN). Reactions were carried out in the presence of NADPH plus NADH (0.4 mM each) for 6 h at 37 °C under an atmosphere of O2/CO2. Products were isolated by column and thin layer chromatography 2°. Reaction rates were related to the amount of protein in the homogenate.
Protein and radioactivity measurements Protein values were estimated by the method of Bradford 2. Radioactivity was measured using either Triton X-toluene-Omnifluor (DuPont NEN) or toluene-Liquifluor (DuPont NEN) scintillation fluid. Counting time was set to yield a dpm S.D. of less than 5%.
0
,E 2.
Right
Loft
Left
Right
TEMPORAL LOBE ANDROGEN RECEPTOR 12.
Male
Female et
¢
10' .e
~E
•
6.
M 0
E
4.
."
.,..
U_
.
2. Right
Left
Right
Loft
Side Fig. 1. Androgen receptor levels in the frontal and temporal lobes of the cerebral cortex. Dashed lines connect right and left side values in individual male and female subjects. Open bars represent the means of the respective groups•
RESULTS
Androgen receptor concentration A R levels in the right frontal lobe of male subjects were higher than in the left frontal lobe (for each and every subject; A N O V A : P < 0.05; Wilcoxon's signed rank test (one tailed): P < 0.05) (Fig. 1), while in female fetuses, there was no consistent pattern in the distribution of the receptor between the two sides of the frontal lobe. In contrast, A R values in the left temporal lobe of male subjects were consistently higher than in the right temporal lobe (Wilcoxon's signed rank test: P < 0.05). As with the frontal lobe, female fetuses exhibited no consistent pattern in the distribution of A R between the two sides of the temporal lobe. Average cortical A R levels in males were: 3.4 (F-Rt), 0.4 (F-Lft), 1.4 (Tmp-Rt) and 3.5 fmoles/mg protein (Tmp-Lft), while in females, concentrations were 1.2 (F-Rt), 1.4 (F-Lft), 4.5 ( T m p - R t ) and 1.1 fmol/mg protein (Tmp-Lft). Substantial variation was noted between subjects in right minus left hemisphere A R levels. For the frontal
124 ,---
c - . _C "-- G)
• Frontal Cortex o Temporal Cortex
0~3_ i-
5
1•0-
•
5 a - R E D U C T A S E ACTIVITY
4.
........................... • -..... ..-"" " ' " ' " ' " ' "•
hE
~ "'"'"..
• ,.-""
0
0
08. Male Female
. -c
~ o
•
0.6-
rJ
0
WE
2
~
o
-0
E
t3._
<
0 3.8
43
48
518
o s'.3
0.4-
~
0
0.2-
0.0 Mus
CFr
CFI
CYr
CYl
Bruin Weight (gm)
Fig. 2. The relation between brain weight in individual male fetuses and right minus left side androgen receptor levels in the frontal lobe (closed circles) and temporal lobe (open circles) of the cerebral cortex• While temporal lobe AR side differences did not appear to be related to brain weight, frontal lobe AR differences could be correlated to brain weight• This relationship was best described by the polynomial equation, Y = -49.7 + 19.9X-1.84X 2, as determined by least squares regression analysis (rE = 0.857)• The dashed line describes the fitted equation.
lobe of male subjects, this variation a p p e a r e d to be related to brain weight (Fig. 2, second degree polynomial, r e = 0.857, regression F2, 2 = 5.99). H o w e v e r , for the male t e m p o r a l lobe and for both lobes in females (data not shown), there was no a p p a r e n t relationship between the right-left side distribution of A R levels and brain weight. Total male and female brain weights were significantly different ( A N O V A : P -- 0.04) averaging 4.93 + 0.25 g for males and 3.93 + 0.34 g for females•
Enzyme levels T h e r e was no statistical difference in aromatase activity b e t w e e n any of the tissues or sexes which were e x a m i n e d including muscle (Fig. 3). M o r e o v e r , in contrast to A R , specific side differences were not found 1.0-
AROMATASE ACTIVITY ~- 0.8._c
[--]
£ 0.6-
Male Female
o_
E 0.40
E 0.2-
13..
oo
rTJ Mus
CFr
CFI
CTr
CTI
Tissue
Fig. 3. Aromatase activity in muscle and cerebral cortex of male and female fetuses. Activities were measured in skeletal muscle (Mus) and in the frontal and temporal lobes of the cortex on the right (CFr, CTr) and left (CFI, CTI) sides• Data represent the means and S.E.M.
Tissue
Fig. 4.5a-Reductase activity in muscle and cerebral cortex of male and female fetuses. Activities were measured in skeletal muscle (Mus) and in the frontal and temporal lobes of the cortex on the right (CFr, CTr) and left (CFI, CTI) sides• Data represent the means and S.E.M.
within individual subjects ( d a t a not shown). For 5a-reductase activity, A N O V A indicated a significant tissue effect ( P < 0.001) which could be explained by higher enzyme activities in the various cortex regions than in skeletal muscle (Fig. 4). T h e r e was no significant sex, regional, side or individual difference in 5a-reductase activity in the cerebral cortex. DISCUSSION
The current study d e m o n s t r a t e s a unilateral distribution of A R in the frontal and t e m p o r a l lobes of the male cerebral cortex. This is the first r e p o r t of s e x - d e p e n d e n t biochemical lateralization of the developing p r i m a t e brain. The fact that neither a r o m a t a s e n o r 5a-reductase exhibited lateralization indicates that the o b s e r v e d distribution of A R is not a general p h e n o m e n o n . A n d r o g e n s affect the d e v e l o p m e n t of neural systems in other species and in conjunction with the observed unilateral distribution of A R in the m o n k e y cerebrum, may influence cortical d e v e l o p m e n t in primates. Evidence in birds and rodents indicates a direct facilitatory effect of androgens on synaptogenesis and an inhibitory effect on synapse elimination 4'12'27. A n d r o g e n s also stimulate the proliferation of glial and endothelial cells 8, retard neuronal loss resulting from a x o t o m y 28 and promote neurite outgrowth from explants 2s. A l t h o u g h several of these actions may be m e d i a t e d through the aromatization of androgens to estrogens, some may d e p e n d directly upon androgens. In the androgensensitive levator ani muscle of the male rat, synapse elimination occurs m o r e slowly following birth than in the androgen-insensitive extensor digitorum longus muscle 12. Specific antiandrogens also have been shown to block the
125 masculinization of the spinal nucleus of the bulbocavernosus 5. The similarity between muscle and cortical aromatase activities, noted in the current study, as well as the variable distribution or absence of cytosolic estrogen receptors on Day 70 (Sholl, unpublished) strengthen the conclusion that any potential effects of androgens on the primate cerebral cortex do not depend upon their aromatization to estrogens. This conclusion is also supported by the observation that administration of dihydrotestosterone, a non-aromatizable androgen, to pregnant monkeys during this early stage of development results in more masculine behavior in female offspring 9. It is not clear how the side differences in AR levels arise, although the apparent correlation in males between these differences in the frontal lobe and total brain weight suggests that, at least in this region, they may arise from the maturational processes which occur around this time. This is supported by preliminary observations from our laboratory on Day 50 fetuses which revealed little or no lateralization of A R regardless of sex, and in several fetuses, levels were undetectable. Although it is possible that side differences in measured A R levels reflect differences in receptor stability and/or degradation, a more plausible explanation is that the receptor is localized within specific cell types or structures which are themselves unequally distributed. If one hypothesizes that there are early side differences in hemisphere maturation, regardless of sex, then the presence of circulating androgen may accentuate the maturational imbalance through interaction with the AR. This might occur as a result of changes in the rate of synapse formation or elimination. A disproportionate stimulation of these processes on one side, as a consequence of higher AR levels on that side, might lead in turn to a decline on the contralateral side. In this regard, it has been proposed that initially there is a pool of neurons which are competitively trying to establish connections with one another 6. If neurons fail to establish these connections, they die. The unilateral distribution of AR may favor the formation of synapses on the side which exhibits higher A R levels. As a result, neurons on the opposite side, which are trying to establish contralateral connections, would tend to decline in numbers. With this loss would come a diminution in AR-associated activity. In contrast, when circulating androgen levels are low, as in the female fetus, the maturation of right and left cerebral hemispheres would tend to proceed in a more random fashion. Evidence in primates which supports the concept of neuronal competition comes from the work of Goldman et al. 7. After unilateral lesioning of the frontal lobe of a fetal monkey ( - E l l 0 ) , the frontal lobe on the opposite
side not only developed connections to the ipsilateral thalamus and caudate nucleus (normal pattern), but a!so established connections with these structures on the side of the lesion. It has been suggested 6 that the compensatory increase in contralateral connections from the non-lesioned side was due to the fact that the lesion removed competing neurons from the pool of those trying to establish connections. Whatever the cause of the unilateral distribution of AR in the male fetus, it still remains that such a distribution pattern was noted. Moreover, this implies that the degree of androgen action in the cortex is dissimilar between the two sides, offering a possible explanation for the ontogeny of sex differences in side-dependent cortical activities and pathologies. This explanation (at least for the frontal lobe) is in line with the hypothesis of Geschwind and Galaburda 6 that there is an androgen-dependent delay in left hemisphere maturation. 5a-Reductase activity was significantly higher in the various cortical regions than in muscle suggesting its importance in these brain areas. This enzyme catalyzes the formation of dihydrotestosterone from testosterone. Since dihydrotestosterone binds more avidly than testosterone to AR, elevated 5a-reductase levels may serve to enhance side-specific, AR-mediated changes. In a previous study, we noted high levels of 5a-reductase in the fetal monkey corpus callosum 2°. On the basis of this finding, it is tempting to speculate that the enzyme helps to concentrate dihydrotestosterone in the corpus callosum. In so doing, the reductase may facilitate androgendependent receptor activity in those cortical regions receiving caUosal projections, particularly in those cells which contain elevated A R levels. When estradiol is implanted unilaterally into the rat hypothalamus, there are differential behavioral and neuroendocrine effects depending upon the side of exposure 14'~8. In addition, cytosolic estrogen receptor levels in the rat cerebral cortex on postnatal day 2 exhibit a significant right-left asymmetry depending o n s e x 19. Although the behavioral and neuroendocrine importance of these phenomena remain to be established, they do suggest that the rat may be an important model for future studies of biochemical asymmetry and the ontogeny of this asymmetry. Recently, Tobet and Fox 24 using a monoclonal antibody, 'AB-2', noted an asymmetrical distribution of immunoreactivity in the neonatal male rat brain in a region caudal to the optic chiasm. A higher density of immunoreactivity was evident in radial glia on the right side. Although females did not exhibit the same degree of asymmetry, early treatment with testosterone shifted the distribution pattern closer to that found in males. This study not only provides additional evidence
126 of asymmetry in the developing rat brain, but also describes an experimental approach which may be applicable to future investigations into the ontogeny of brain asymmetry in both the rat and primate.
REFERENCES 1 Barr, W.B., Jaffe, J., Wasserstein, J., Michelson, W.J. and Stein, B.M., Regional distribution of cerebral arteriovenous malformations, Arch. Neurol., 46 (1988) 410-412. 2 Bradford, M., A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of proteindye binding, Anal. Biochem., 72 (1976) 248-254. 3 Clark, A.S., MacLusky, N.J. and Goldman-Rakic, P.S., Androgen binding and metabolism in the cerebral cortex of the developing rhesus monkey, Endocrinology, 123 (1988) 932-940. 4 DeVoogd, T.J., Nixdorf, B. and Nottebohm, E, Synaptogenesis and changes in synaptic morphology related to acquisition of a new behavior, Brain Research, 329 (1985) 304-308. 5 Fishman, R.B. and Breedlove, S.M., Androgen blockade of bulbocavernosus muscle inhibits testosterone-dependent masculinization of spinal motoneurons in newborn female rats, Soc. Neurosci. Abstr., 13 (1987) 1520. 6 Geschwind, N. and Galaburda, A.M., Cerebral Lateralization, Biological Mechanisms, Associations, and Pathology, MIT, Cambridge, 1987, 283 pp. 7 Goldman, P.S. and Galkin, T.W., Prenatal removal of frontal association cortex in the fetal rhesus monkey: anatomical and functional consequences in postnatal life, Brain Research, 152 (1978) 451-485. 8 Goldman, S.A. and Nottebohm, E, Neuronal production, migration and differentiation in a vocal control nucleus of the adult female canary brain, Proc. Natl. Acad. Sci. U.S.A., 80 (1983) 2390-2394. 9 Goy, R.W., Differentiation of male social traits in female rhesus macaques by prenatal treatment with androgens: variation in type of androgen, duration, and timing of treatment. In M.J. Novy and J.A. Resko (Eds.), Fetal Endocrinology, Academic Press, New York, 1981, pp. 319-339. 10 Goy, R.W., Uno, H. and Sholl, S.A., Psychological and anatomical consequences of prenatal exposure to androgens in female rhesus. In T. Mori and H. Nagasawa (Eds.), Toxicity of Hormones in Perinatal Life, CRC Press, Boca Raton, 1988, pp. 127-144. 11 Handa, R.J., Connolly, P.B. and Resko, J.A., Ontogeny of cytosolic androgen receptors in the brain of the fetal rhesus monkey, Endocrinology, 122 (1988) 1890-1896. 12 Jordan, C.L., Letinsky, M.S. and Arnold, A.P., Synapse elimination occurs late in the hormone-sensitive levator ani muscle of the rat, J. Neurobiol., 19 (1988) 335-356. 13 Maccoby, E.M. and Jacklin, C.N., The Psychology of Sex Differences, Stanford University Press, Stanford, 1974, 634 pp. 14 Nordeen, E.J. and Yahr, P., Hemispheric asymmetries in the
Acknowledgements. The work described in this article, Publication 29-030 of the Wisconsin Regional Primate Research Center, was supported by NIH Grants HD18865 and RR00167. All experiments presented in this manuscript were performed by following the standards established by the Animal Welfare Act and by documents entitled 'Principles for the Use of Animals' and 'Guide for the Care and Use of Laboratory Animals'.
behavioral and hormonal effects of sexually differentiating mammalian brain, Science, 218 (1982) 391-394. 15 Pomerantz, S.M., Fox, T.O., Sholl, S.A., Vito, C.C. and Goy, R.W., Androgen and estrogen receptors in fetal rhesus monkey brain and anterior pituitary, Endocrinology, 116 (1985) 83-89. 16 Pomerantz, S.M. and Sholl, S.A., Analysis of sex and regional differences in androgen receptors in fetal rhesus monkey brain, Dev. Brain Res., 36 (1987) 151-154. 17 Resko, J.A., Ellinwood, W.E., Pasztor, L.M. and Buhl, A.E., Sex steroids in the umbilical circulation of fetal rhesus monkeys from the time of gonadal differentiation, J. Clin. Endocrinol. Metab., 50 (1980) 900-905. 18 Roy, E.J. and Lynn, E.M., Asymmetry in responsiveness of the hypothalamus of the female rat to estradiol, Physiol. Behav., 40 (1987) 267-269. 19 Sandhu, S., Cook, P. and Diamond, M.C., Rat cerebral cortical estrogen receptors: male-female, right-left, Exp. Neurol., 92 (1986) 186-196. 20 ShoU, S.A., Goy, R.W. and Kim, K.L., 5a-Reductase, aromatase, and androgen receptor levels in the monkey brain during fetal development, Endocrinology, 124 (1989) 627-634. 21 Sholl, S.A. and Pomerantz, S.M., Androgen receptors in the cerebral cortex of fetal female rhesus monkeys, Endocrinology, 119 (1986) 1625-1631. 22 Taylor, D.C., The influence of sexual differentiation on growth, development, and disease. In J.A. Davis and J. Dobbing (Eds.), Scientific Foundations of Paediatrics, Saunders, Philadelphia, 1974, pp. 29-44. 23 Taylor, D.C. and Ounsted, C., Biological mechanisms influencing the outcome of seizures in response to fever, Epilepsia, 12 (1971) 33-45. 24 Tobet, S.A. and Fox, T.O., Sex- and hormone-dependent antigen immunoreactivity in developing rat hypothalamus, Proc. Natl. Acad. Sci. U.S.A., 86 (1989) 382-386. 25 Toran-Allerand, C.D., Sex steroids and the development of the newborn mouse hypothalamus and preoptic area in vitro: implications for sexual differentiation, Brain Research, 106 (1976) 407-412. 26 Wittig, M.A. and Petersen, A.C., Sex-Related Differences in Cognitive Functioning, Academic Press, New York, 1979, 378 PP. 27 Wright, L.L. and Smolen, A.J., Neonatal testosterone treatment increases neuron and synapse numbers in male rat superior cervical ganglion, Dev. Brain Res., 8 (1983) 145-153. 28 Yu, W.A., Sex difference in neuronal loss induced by axotomy in the rat brain stem motor nuclei, Exp. Neurol., 102 (1988) 230-235.